field test unit no 001 "model T"

[Aziz]

Fully differential and DC accurate ... max input is +/- 2.5 volts to the ADC channels ..

BRAVO!!!! I like it.
Forget my projects. The QED2 will be a killer-app. That's a game changer.
(And I can give you a lot of high-tech info as well (WBGB, noise elimination, etc..))
Aziz

[Aziz]

He is sampling at 2.5 MHz according to the FFT spectrum.

[moodz]

And I hope, you drive the ADC-chip in a fully differential mode (differential coil would make sense). That would give a superior SNR.

And I think, you could implement some advanced techniques to reduce further noise. Make sure, your platform has enough RAM memory and CPU/FPU processing power.

Aziz


PS: Get ready to get legally sued to be better than <YKW>.
*LOL*

Fully differential and DC accurate ... max input is +/- 2.5 volts to the ADC channels ..

[chudster]

..been working ..away at the signal chain ... happily I can now say the descrimination and target channels are working at 24 bit sampling with a bandwidth exceeding 1 Mhz and direct sampling of the frontend. Noise floor is down at -140db .. and I have not even applied the SAT yet ... I can extend the bandwidth to 10 Mhz but the effective bits will drop to 16 ..... not bad for sub $100 hardware .

below are spectrum pix of the target and descrim channels .... I cant show waveforms with target and descrim infos as that may give away too much at this "sensitive" time .....the software PI is coming .... and its a race to the bottom.

[ATTACH]21352[/ATTACH]
[ATTACH]21353[/ATTACH]

Moodz,

Are you saying that you are sampling 24 bits at 2MSPS to get a bandwidth of 1 MHz or is this of the demodded signal at a lower sample rate?

Chudster

[Aziz]

And I hope, you drive the ADC-chip in a fully differential mode (differential coil would make sense). That would give a superior SNR.

And I think, you could implement some advanced techniques to reduce further noise. Make sure, your platform has enough RAM memory and CPU/FPU processing power.

Aziz


PS: Get ready to get legally sued to be better than <YKW>.
*LOL*

[Aziz]

Hey Moodz,

looks like you are No 1 candidate to get the WBGB technology. 1 MHz bandwidth is by far enough.

Direct sampling & decoding is the way to go. Get a high number cruncher for this purpose (DSP).
Cheers,
Aziz

PS: Ooops, forgot something:
It's the end of <you know who>!!!

[moodz]

....been working away at the signal chain ... happily I can now say the descrimination and target channels are working at 24 bit sampling with a bandwidth exceeding 1 Mhz and direct sampling of the frontend. Noise floor is down at -140db .. and I have not even applied the SAT yet ... I can extend the bandwidth to 10 Mhz but the effective bits will drop to 16 ..... not bad for sub $100 hardware .

below are spectrum pix of the target and descrim channels .... I cant show waveforms with target and descrim infos as that may give away too much at this "sensitive" time .....the software PI is coming .... and its a race to the bottom.

[hdphilip]

now i'll have to see if it works on the bench.....to know if it's solved

[PATCHES JUNIOR]

Anyone solved the riddle?

Nah, we all gave up. What is the answer?

[Davor]

Nah. I think I'll just wait to see what that is. So far I'm not impressed by the whole PI thing at all, and the only development I can foresee some tangible success is the Tinkerer's rig.

[Aziz]

Anyone solved the riddle?

[Davor]

It seem to me as if there is no transformation of a damping resistance to a target, hence decay is prolonged. Flyback, my ...

This very reason - resistance transformation - gave me impression that I'd get much more by step voltage excitation. After half a year I still think I'd get much more by step voltage excitation. Losses with PI resistor are just too great.

[sido]

I am thinking towards a matched impedance step up transformer to the mono M/L coil in the front end??

Sid

[hdphilip]

chudster

Maybe hdphilip you meant with Moodz' circuit and resistive damping (if that is possible) although Moodz says he has separate real and reactive elements with the load switched off near the 'undershoot' so maybe you are asking for the switch off not to happen?

i don't think moodz is switching anything, i mean, by the way of a mosfet, i suppose you could, let's review his clues....

here, i'll pm you my ideas, and see if the make sense with your ideas...

Philip

[chudster]

Its a sim as I am not near the bench .... the params are for a 18 inch minelab mono coil.


PS ....

we dont want any naked developers ... this is a clothes on thread however you are permitted to run around screaming and pulling your hair out.

moodz

[ATTACH]21050[/ATTACH]

So interpretation is that there is no operation of your circuit. Flyback time is greatly extended. Decay is (critically damped??) from the resistor only. If so that is as expected.

Maybe hdphilip you meant with Moodz' circuit and resistive damping (if that is possible) although Moodz says he has separate real and reactive elements with the load switched off near the 'undershoot' so maybe you are asking for the switch off not to happen?

Chudster